A Stochastic Model for Generating Synthetic Daily Household Electricity Demand Profiles from Smart-Meter Data - Validation on the Optimal Sizing of Batteries
ACM SIGEnergy Energy Informatics Review · 01.09.2026 · 6 (3) · 32-44
DOI: 10.1145/3861130.3861133 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM SIGEnergy Energy Informatics Review, online erschienen am 11.11.2026. Verfasst von Plouton Grammatikos und 2 weiteren Autor:innen. Die Arbeit nennt 30 Literaturangaben.
Abstract (Originalsprache)
A data-driven stochastic model is proposed to generate synthetic daily household electricity demand scenarios for a distribution grid. The method clusters households by their average demand profiles to identify distinct consumption patterns and fits a multivariate Gaussian mixture model (GMM) to the logarithm of the demand residuals relative to each cluster mean. This formulation accurately captures both the average and median demand profiles of the original dataset across different days of the week and periods of the year. The model's ability to reproduce key statistical properties of the original data is evaluated with various key performance metrics, with particular emphasis on the joint distribution of daily peak power and energy consumption, which governs the coupling between the grid's power and energy needs. To demonstrate the practical value of the method, an optimal siting and sizing problem for a battery energy storage system (BESS) in a distribution network is formulated, and the results obtained using the synthetic scenarios are compared against average demand profiles reported in the literature. Because the peak-power-to-energy ratio is accurately modeled, the sizing algorithm computes 3.5 times less energy capacity for the considered use case, requiring 40% less investment to meet the same power requirements. The proposed method is of direct interest to distribution system operators (DSOs) for evaluating both wire and non-wire alternatives for network reinforcement.
Zitieren
APA
Grammatikos, P., Ali, A. M., & Sossan, F. (2026). A Stochastic Model for Generating Synthetic Daily Household Electricity Demand Profiles from Smart-Meter Data - Validation on the Optimal Sizing of Batteries. ACM SIGEnergy Energy Informatics Review, 6(3), 32-44. https://doi.org/10.1145/3861130.3861133
Vancouver
Grammatikos P, Ali AM, Sossan F. A Stochastic Model for Generating Synthetic Daily Household Electricity Demand Profiles from Smart-Meter Data - Validation on the Optimal Sizing of Batteries. ACM SIGEnergy Energy Informatics Review. 2026;6(3):32-44. doi:10.1145/3861130.3861133
BibTeX
@article{Grammatikos2026,
author = {Grammatikos, Plouton and Ali, Ali Mohamed and Sossan, Fabrizio},
title = {A Stochastic Model for Generating Synthetic Daily Household Electricity Demand Profiles from Smart-Meter Data - Validation on the Optimal Sizing of Batteries},
journal = {ACM SIGEnergy Energy Informatics Review},
year = {2026},
volume = {6},
number = {3},
pages = {32--44},
doi = {10.1145/3861130.3861133}
}
EUVIMED